Fichas
association rules
Fichas del conjunto (10)
- 1supportprobability of randomly picked item will contain the antecedant and consequent
- 2Confidenceconditional probability of random pick will contain antecedant and consequent given that it will contain antecedant
- 3Liftratio of confidence to expected confidence;
expected confidence = probability of the consequent
A lift ratio larger than 1.0 implies that the relationship between the antecedent and the consequent is more significant than would be expected if the two sets were independent. The larger the lift ratio, the more significant the association. - 4what does confidence and lift mean?high lift (>1) and confidence (>90%) indicated a good association
- 5Collaborative filteringbased on notion of similarity:
-user based
-item based
-content based
not only predicts who will like but how much as well - 6measure of collaborative filteringeuclidean distance
cosine similarity
pearson coefficient
jacard coefficient - 7euclidean distancesmaller the distance better the similarity;
issue with approach is users may not have bought items in common (sparsity) still the distance might be small - 8cosine similarity1 => perfect
0 => dissimilar
-1 => perfect negative similarity - 9pearson coefficient1 => perfect similarity
0 => no similarity
-1 => negative correlation - 10Jacard coefficientjacard coeff. = Nc / (Na+Nb -Nc)
0 => no correlation
1=> perfect correlation
Información
- Descripción
- asd Business Analytics Fichas sobre association rules, creado por Mukundhan Srinivasan el 31/05/2016.
- Creado por
- Mukundhan Srinivasan
- Visitas
- 0
- Valoración